© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 22361
    Prev
    Next
    INDEX
    Explanations

    definitions or instances of what constitutes something

    oai_token-act-pair · gpt-3.5-turbo

    the word "constitute" and its variations, indicating definitions or descriptions of what something is or represents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Ĭ±
    -0.67
    Patch
    -0.63
     smiles
    -0.61
     Hearts
    -0.59
     smarter
    -0.59
    tun
    -0.58
     XP
    -0.58
    mo
    -0.58
     Wicked
    -0.57
     Ammo
    -0.56
    POSITIVE LOGITS
     constitutes
    3.25
     constitute
    2.90
     constituted
    2.75
     comprise
    1.81
     comprises
    1.77
     amounted
    1.77
     qualifies
    1.70
     constit
    1.43
     entails
    1.39
     comprising
    1.27
    Activations Density 0.015%

    No Known Activations